[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <13521f9f5ebe1cd0a85bf31b5ad5cba0@codeaurora.org>
Date: Mon, 19 Sep 2016 14:41:10 -0400
From: bdegraaf@...eaurora.org
To: Laura Abbott <labbott@...hat.com>
Cc: catalin.marinas@....com, will.deacon@....com, james.morse@....com,
mark.rutland@....com, andre.przywara@....com,
jungseoklee85@...il.com, apinski@...ium.com,
linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
timur@...eaurora.org, cov@...eaurora.org
Subject: Re: [RFC] arm64: Ensure proper addressing for ldnp/stnp
On 2016-09-19 14:28, Laura Abbott wrote:
> On 09/19/2016 10:36 AM, Brent DeGraaf wrote:
>> According to section 6.3.8 of the ARM Programmer's Guide, non-temporal
>> loads and stores do not verify that address dependency is met between
>> a
>> load of an address to a register and a subsequent non-temporal load or
>> store using that address on the executing PE. Therefore, context
>> switch
>> code and subroutine calls that use non-temporally accessed addresses
>> as
>> parameters that might depend on a load of an address into an argument
>> register must ensure that ordering requirements are met by introducing
>> a barrier prior to the successive non-temporal access. Add
>> appropriate
>> barriers whereever this specific situation comes into play.
>>
>
> Was this found by code inspection or is there a (public) exciting test
> case to observe this behavior?
>
> Thanks,
> Laura
>
Code inspection only.
Brent
Powered by blists - more mailing lists